Do political directors affect the financial performance in
the UK?
Master thesis, MSc Accountancy
University of Groningen, Faculty of Economics and Business January 20th, 2020 WALTER MEISSNER S3858871 E-mail: w.g.meissner@student.rug.nl Word count: 8546 Supervisor university: Dr. S. Mukherjee Supervisor field of study:
1
ABSTRACT
In this study I investigate if former politicians, who are now corporate directors, affect financial performance of firms in the United Kingdom (UK). I argue that, based on the resource dependency theory, political directors are unlikely to provide usable resources to firms since the UK has a mature economy with an efficient bureaucracy and little corruption. Resource dependency theory suggests that network and knowledge about the governments adds value for the firms. Prior research finds ambiguous relation between political directors and firm performance in non-UK countries. Moreover, they did not distinguish among the political professions of the former politicians, while these
differences are likely to bring different resources to the firm. Therefore, I investigate not only the association between political directors and financial performance of UK firms, but also of senior bureaucrat and minister directors. I expect that political directors are associated with negative financial performance in the UK. Lastly, I investigate the effect of successful firms on the
beforementioned relations. I expect that successful firms have a moderating effect, since it seems that they have more capabilities for making the right choices in appointing a director. My findings are consistent with my predictions. This study contributes to the resource dependency theory by showing under what conditions political directors are unlikely to positively affect firm performance.
2
Contents
1. INTRODUCTION 3
2. THEORETIC FRAMEWORK & HYPOTHESES 5
3. RESEARCH DESIGN 9
3.1. Sample and Data 9
3.2. Variables 9 3.2.1 Dependent Variables 9 3.2.2 Explanatory Variables 9 3.3. Empirical model 11 4. RESULTS 12 4.1. Descriptive statistics 12
4.2. Person Correlations Coefficient 12
4.3. Regression 12
4.4. Robustness test 14
5. CONCLUSION 15
6. Literature 16
3
1.
INTRODUCTION
Government policy, regulation, and enforcement are major forces in the external environment of the firms (Hillman & Hitt, 1999). The dependency on governments creates risk and uncertainty. To mitigate this uncertainty, resource dependency theory suggests that firms employ political directors to connect with governments. Politically connected firms can protect and advance the economic interests of the firms (Hillman & Hitt, 1999). Political directors are expected to bring their knowledge and experience of the government to the firm. Prior research shows that political directors influence the financial performance of the firm (Hadani & Schuler, 2012; Boubakri et al., 2008). Especially in less developed countries with a high degree of corruption, political connection is beneficial for financial performance (Faccio, 2010). But is this useful in developed countries, especially in the UK? In this thesis I examine if the influence of political directors, especially senior bureaucrats, affects financial performance in the UK. In addition, I will examine the moderating role of the successful firms in this relation.
Directors have an important position in the firm. They give advice to managers and monitor the firm’s corporate governance (Adams & Ferreira, 2007). This position is very influential for the firm’s decision-making. There are a few board positions in each firm, making it important to choose the best director for the job. Following resource dependency theory, firms can benefit from the network and the political connections political directors could bring. Prior research shows that in countries with more corruption, political connections are beneficial for firm financial performance (Faccio, 2010). Prior research about the relationship between political directors and financial performance have different outcomes. According to Hadani & Schuler (2012), firms of S&P 1500 placing former politicians in the boards experience inferior market performance and accounting performance than firms without politically connected directors. Boubakri et al. (2008) investigates firm political connections in 41 countries, separated into developed or developing countries. They too find a negative relation between political directors and accounting performance for firms. But in Germany, politically connected firms have better accounting performance and market performance than non-connected firms (Niessen & Ruenzi, 2009). My study builds upon previous research in several ways. First, I specifically focus on political directors in the UK. Second, I distinguish several types of “political board members”, senior bureaucrats being one of them. This group has not been investigated yet, while this group brings other resources to the firm than, for example, ministers. Moreover, senior bureaucrat directors are the most prevalent type of political directors in the UK. Third, we assess the moderating role of firm success on political director’s impact on financial performance.
The resource dependency theory offers an explanation of why firms appoint political directors to their boards. According to the resource dependency theory, directors are expected to bring resources in the
4
form of expertise and experience into the firm. Appointing political directors ensures resources to reduce the risk and uncertainty caused by the dependency of the government. Political directors create channels of access for the firm to the government, to influence policy formulation and enforcement (Pfeffer & Salancik, 1978; Hillman, 2005). Further, they provide knowledge about the inner workings of the governments (Hillman, 2005). The UK is a mature economy with a developed market. Therefore, I posit that political ties in the UK is not beneficial for firm’s financial
performance.
Political directors can be divided into former senior bureaucrats and ministers. Both bring their own experiences and resource into the firm. Ministers represents the highest position in government but serve only one government period and therefore have a short tenure in policy-making roles. Whereas, senior bureaucrats are not attached to a government cycle. In the UK, senior bureaucrats hold their high position after the change of governments. According to the existing political literature, senior bureaucrats have expertise and private information. This leads to agency problems between ministers and senior bureaucrats (Laffont & Martimort, 2002:3). Further, the information advantages lead to selective advising by senior bureaucrats to ministers and shape their political agenda and policy preferences (Blom-Hansen et al., 2017). Senior bureaucrats are likely to bring more resources into the firm than ministers, but the usefulness of political experience in the UK is likely not beneficial.
Therefore, I posit that senior bureaucrat directors have negative association with financial performance of the firm.
Directors of successful firms can take better decision that results in better performance (Ireland & Miller, 2004). The board is more able to attract better board directors to the firm. Therefore, I expect that firm success moderates the negative effect of politically connected directors in the board and financial performance. Further, I expect that this moderates the negative effect for political directors and for senior bureaucrat directors.
To assess my propositions, I use UK data in this thesis. Specifically, I use hand-collected data on the background and work experience of political directors in the government. Building on our hand-collected data I was able to subcategorise the political directors as ministers, senior bureaucrats, advisors and military. In this thesis I specifically zoom in to ministers and senior bureaucrats, because these backgrounds are the most common for political directors. Military is excluded because they have a rather different background than the other categories.
I found that political directors are likely to have a negative association with firm performance. By disaggregating the political background into senior bureaucrats and ministers, I found that senior bureaucrat directors are likely to have a negative association with financial performance. Finally, I
5
found that successful firms moderate the negative effects between political directors and senior bureaucrat directors with financial performance.
My thesis contributes to the resource dependency theory by showing that political directors in the UK are likely to have a negative association to financial performance. Since the UK is a developed country and has low corruption in the bureaucracy. Furthermore, the economy and the corporate governance regulations are mature. Related studies were not focused on primary accounting performance and do not have recent data. Hadani & Schuler (2012) investigated a US sample from 1998 till 2008 and focused on Corporate Political Investment and market performance. Boubakri et al. (2008)
investigated the political connections and financial performance on a multi-country level from 1980 till 2002. They distinguished the following political connections: a member of parliament, minister or other top appointed-bureaucrat. Niessen & Ruenzi (2009) investigated from which political party, political directors were part of and its relation to financial performance.
I also contribute to the resource dependency theory by showing that the resources senior bureaucrats bring into firms in the UK are likely to have negatively associated with financial performance. This seems to be related to the characteristic of the UK. Since the UK has a mature economy with an efficient bureaucracy and little corruption. To the best of my knowledge, this is the first study on the link between political connections and firm performance that distinguishes senior bureaucrat directors and minister directors. Existing literature treats directors with different political
backgrounds equally, while they can come with substantial implications for the resources they bring to the firm and, subsequently, firm performance. In the UK, firms mostly appoint former senior bureaucrats, while they have negative effects on financial performance. This study thereby shows that appointing political directors has a negative side.
The last contribution that we investigate the relation between political directors and performance specifically for successful firms. To the best of my knowledge, this is the first study that does this. The findings are in line with the theory: successful firms can appoint more qualified directors, which moderates the negative effect of political directors on performance.
The remainder of this thesis is structured as follows. In the second section I review the literature and the hypothesis development. The third section discusses the methodology that is used in this study. The fourth section presents the results. The last section presents the conclusions.
2.
THEORETIC FRAMEWORK & HYPOTHESES
Following the resource dependency theory, politically connected directors are expected to bring external resources, such as access to policy formulation and enforcement (Pfeffer & Salancik, 1978; Hillman, 2005). Furthermore, they are expected to supply firms with private knowledge of the inner
6
workings of governments (Hillman, 2005). Political connections are expected to bring advantages over non-connected firms. For example, firms with political connections are more likely to be bailed out (Faccio et al., 2006) and can borrow more with lower interest rates (Fisman, 2001; Sapienza, 2004). On the other hand, the competence of the politicians bring as outside directors may be of limited value for the company.
There are two main functions of the board: decision management, i.e. initiation and implementation of decisions, and decision control, i.e. ratification and monitoring of decisions (Fama & Jensen, 1983). Outside directors do a better job of monitoring management (Helland & Sykuta, 2003). Political directors are more likely outside directors in the board (Agrawal & Knoeber, 2001). Kor & Sundaramurthy (2008) claim that politicians do not supervise and discipline management with the same level of competence and vigilance as outside directors with greater business acumen and commercial experience, such as former CEOs and industry insiders. According to Shaukat et al. (2018) there is a positive relation between board governance and firm performance, and better monitoring is associated with better performance. The question that arises is whether the advantages in the UK of having politically connected directors outweigh the disadvantages (or vice versa). The resources that political directors expect to bring in firms are more beneficial in less developed countries with a high degree of corruption (Faccio, 2010). The UK is a developed country and has low corruption in the bureaucracy (Dunsire, 1988). The economy and the corporate governance
regulations are mature (Mitchell, 2009; Thomsen, 2003). The incentive of the CEO is market-driven as their compensation is based on firm performance (Ozkan, 2011). These characteristics of the UK ensures more disadvantages to add a political director to the board. So, in the UK firms are unlikely to benefit by employing a political director. Further, the appointing of a political director holds a more qualified director from that position in such developed economy.
Previous research about the relation between political directors in the board and financial
performance focused on multi-country study, Germany and the United States. This thesis investigates this relation for firms in the UK. Though our aim is not to provide a comprehensive overview, it is clear that differences and similarities between these countries exist. A key similarity is that the board of directors has significant power to guide the strategy and monitor the officers of the company. Also, the governance systems of the UK and US share similarities. Both countries have high protection of minority rights and common law systems (La Porta et al., 1998). Differences exist in corporate governance regulations. In the UK, compliance with regulations are voluntary and operate on a comply-or-explain basis (Garrett, 2019; Guest, 2008). Other differences are the enforcement of directors’ legal duties and the role of institutional investors (Guest, 2008). Due to these differences, evidence obtained from US based studies cannot be generalized to the UK. Studying the UK thus contributes to the existing literature.
7
Previous studies use various ways to quantify financial performance contribution of political directors. The measure used in this thesis is accounting performance, which is a commonly used measure for performance evaluation and compensation (Otley, 1999). It is also the most conservative way to measure firm performance. Previous studies present mixed results on this measure. Hadani & Schuler (2012) find that US firms who appoint former officials worsen accounting performance and market performance. Based on an international study, Boubakri et al. (2008) suggest that politically connected firms exhibit poor accounting performance compared to non-connected firms. Niessen & Ruenzi (2009), on the other hand, found that politically connected firms provide better accounting performance in Germany. Politicians in the board also impact firms’ governance. Aggarwal et al. (2012) show that governance quality is weaker when firms have politically connected board members. The resource dependency theory suggests that political directors are appointed to create linkage with government for mitigating the dependency of the government. The UK is a developed country with low corruption. Further, the monitoring role of the political directors is worse than other outside director, while this has influence on firm performance. Because of these reasons, I expect that political connected directors in UK firms are negatively related to firm performance.
Hypothesis 1: Politically connected firms are negative associated with financial performance. Most research on politically connected board member focuses on ministers and other high-ranking positions. However, politically connected board members have a variety of functions in government. The different types (ministers, senior bureaucrats, advisors and military) have different networks, expertise and knowledge (Ruigrok et al., 2006). In this thesis I split politicians in senior bureaucrats and ministers. I disregard advisors and military because senior bureaucrats and ministers are mostly employed political directors, and the military is not comparable with the other types of occupation. The function of a minister represents the highest position in a government. This suggests that they had more influence in politics than senior bureaucrats and advisors. Ministers serve during one government period. Hence, they have a shorter tenure than other functions. The prior experience of ministers is often in business and not in politics (Beckman, 2006).
Senior bureaucrats are different from ministers. Employment of senior bureaucrats is not attached to a government period, so their tenure is longer than that of ministers. Especially in the UK, senior bureaucrats have an important role in advising the ministers. A new government works with the same bench of senior bureaucrats as the former government. The advantage of this system is that
government can start immediately after the elections. In the US, it takes months install a new
administration (Wilson & Barker, 2003). For that reason, in the UK, senior bureaucrats have a longer tenure.
8
According to prior political literature, the political relationship between political authorities and senior bureaucrats is one of principal and agent (Moe, 2005). Senior bureaucrats (agents) have expertise and private information. This private information leads to moral hazard and adverse selection problems (Laffont & Martimort, 2002:3). Through the information advantage senior bureaucrats advise ministers, senior bureaucrats may give the information they want to share and prefer. Therefore, senior bureaucrats can determine the political agenda and shape political preferences regarding issues of the agenda (Blom-Hansen et al., 2017). According to Wood (1988) elected institutions have little influence when bureaucratic resources are substantial. Senior bureaucrats can move outputs completely opposite from what the politicians want.
Concluding, according to the resource dependency theory, senior bureaucrats in the UK have more political resources bring into the firm than ministers. Senior bureaucrats have longer experience in the government than ministers. Because, ministers are in rotation governments while bureaucrat’s tenure is not bound to election periods. Further, senior bureaucrats have an advantage toward
ministers, and can shape their preferences and determine their agenda. The country characteristics of the UK ensures less value of the experience in government is for firms. Because in less developed countries and in countries with high corruption, companies can benefit more from the political connections (Faccio, 2010). Therefore, usefulness of political experience can be questioned. Further, firms change more often to fit in the environment, this is not for governments because there is no need for change. Hence, I hypothesize that more political resources of directors are negative associated with the financial performance of the firm.
Hypothesis 2: Senior bureaucrat directors are negative associated with performance
The effect hypothesized above may differ across (types) of companies. In this thesis, I will specifically study the impact of being a successful firm in the UK on this effect. Board members from these firms have made decisions that resulted in high firm performance and success (Ireland & Miller, 2004). Successful firms hence seem to have more capabilities to make the right choices in uncertain situations. So, I expect the board is more likely to choose politically connected candidates that
positively impact performance. Hence, I hypothesize that appointed politically connected directors in successful firms have a less negative impact on performance.
Hypothesis 3: Political directors in successful firms moderates the negative relationship with financial performance.
9
3.
RESEARCH DESIGN
3.1. Sample and Data
The data used for this thesis is collected from Worldscope and BoardEx. Data of financial statements is collected from Worldscope. BoardEx contains information about the characteristics of directors and firms. For example, gender, tenure, the number of other directorships and retirement status. To use both data source, I matched the data, from Worldscope and BoardEx, on the unique firm-year number.
Lastly, I used unique hand-collected data about the specific background of the political directors. BoardEx has some incomplete background information, I have checked the reliability and completed the political background of directors for several countries. I used Bloomberg and Marketscreener to check the background of each individual director. When I found government functions, I categorized the function into four groups; minister, senior bureaucrat, advisor or military. When I found overlap in functions, I categorized the directors into the highest category. For example, when directors had ministerial and bureaucratic functions, I categorized them into ‘minister’.
I used United Kingdom data focusing on the period from 2002 till 2015. Table 1 shows the
distribution of political connected directors per year. The sample focused only the financial non-utilities and excluded the other categories.
[insert table 1]
3.2. Variables
3.2.1 Dependent Variables
The retrospective measure for performance I used is the accounting performance. Based on previous research I used the Sales per Asset (SALES_PER_ASSET) and the ROA (ROA) as measures. The Sales per Asset is a ratio of the total sales per total assets. This measure is also used in other studies like, Su & Fung (2013). The ROA is a commonly used measure (Boubakri et al., 2008; Su & Fung, 2013), this includes the sales and the costs per asset. Therefore, the ROA gives more detailed information than the Sales per Asset.
3.2.2 Explanatory Variables
In this research the explanatory variables are political directors and directors specified by their political background; former minister, senior bureaucrat and advisors. They must only be appointed in the board of directors. Previous research about this subject for example, Faccio (2006) used a different definition of political connected board members. In that paper, a company is political connected when a politician is one of the top officers is: a member of parliament (MP), a minister or
10
the head of state, or closely related to a top official. This thesis has unique hand-collected data to investigate this relation. The criteria for the separation states in paragraph 3.1. This is the first thesis that makes a separation between minister and senior bureaucrats in the board of directors. The measure of politicians, senior bureaucrats and ministers in the board of directors is in percentages. This gives clearer explanation about the results about the implications of the percentage present political connected directors.
The moderating variable is ‘successful firms’ measured by the TOBINS_Q. According to Su & Fung (2013) this can be used to determine the successfulness of the firm. This variable is calculated as the ratio of the market value of the total assets divided by the book value of the total assets. The number of shares is multiplied by the market price plus the book value of total debt and total assets. The explanatory variable is a contribution to the existing literature if politicians in boards add more value in successful firms than in not successful firms.
Control variables:
For the control variables I used accounting performance and board characteristic variables.
The dependent variables are influenced more variables than the independent variables, according to Herciu & Ogrean (2017) the liabilities (LEVERAGE*) have a significant also influence on the ROA. The profitability of the firm can be increased by an optimal structure of liabilities and equity. Furthermore, the investments in research and development (R&D / TOTAL ASSETS*) are mostly positively related to performance measured by the ROA (Sher & Yang, 2005). The loss dummy for the ROA is used to control the negative ROA in the sample. Also, capital expenditure (CAPITAL
EXPENDITURES/ TOTAL ASSETS*) influence the dependent variables. Because it influences the trust of investors and stakeholders in the company and therefore influence the performance of the firm (Chu, 2019). Closely held shares (CLOSELY_HELD_SHARES*) is used to control the structure of firm’s ownership, because this influence the financial performance of the firm. Also, foreign assets (FOREIGN ASSETS / TOTAL ASSETS*) influence the firm performance according to Chari et al. (2009). This research measures the stock price of firms after an acquisition, when US firm acquire an emerging market firm. The ROA of both firm increases.
Board variables:
There are many board characteristics that also influence the dependent variables. According to Guest (2009) board size (BOARD_SIZE*) has a strong negative impact on profitability of the firm.
Especially for large firms is the relationship between board size and profitability negative. The relation between the proportion of non-executive (BOARD_INDEPENDENCE*) directors and firm performance is according to Mura (2007) positive. It suggests that independent executives can better monitor the firm. Female board members (FEMALE_DIRECTOR*) are positively associated with
11
performance according to Chen et al. (2018). First, firms with female board member invest more in innovations and obtain more patents and have more R&D expenditures. Furthermore, female directors increase monitoring. Further, the proportion of outside directors in the board, due to their presumed independence, theoretically lead to better firm performance (Jensen & Meckling, 1976; Shleifer & Vishny, 1997). Also, the board tenure (BOARD_TENURE*) has influence on the
accounting performance. In Huang & Hilary, (2018) states that increasing board tenure increases the ROA. Lastly, I used the Average quoted board affiliations (QUOTED_BOARD_AFFILIATIONS*) to control the total tenure of the directors.
3.3. Empirical model
This research uses data from 2002 to 2015 of firms from the UK and makes use of panel data. I use a fixed-effect regression to test the hypotheses. This model does not remove only the effect of the treatments, it removes the effects of unobserved confounders that do not vary over time. With this model I control year effects and firm fixed-effects. Further, this fixed-effect regression is a difference-in-difference model, because the value per β is ‘‘one’’ only in the year there is a political director, and ‘‘zero’’ when there is no political director in the board (Fauver et al., 2017). Therefore, β1 and β2 acts as the ‘treatment * post’ variables which is standard in a difference-in-difference model. The three hypotheses are explained in one generic model. All the hypotheses use the dependent variable (ROA and the SALES_PER_ASSET) and the independent variables differs between political
(PERCENTAGE_POLITICAL_DIRECTORS, PERCENTAGE_MINISTER_DIRECTORS and PERCENTAGE_BUREAUCRAT_DIRECTORS). All control variables are the same in the models. Generic model for the direct (table 5) relations
ROA it and SALES_PER_ASSET = β0+ β1(Political Director) + β2(Political Director× Tobin’s Q) + β CONTROLS it +βnFE + ε i,t
The coefficients measure the firm performance of firm i in year t. β1 represents the firms that have political directors, minister director and the senior bureaucrat directors in their board. β2 represents for the moderating effect of the treatment group and the post group. β CONTROLS it represents the control variables of table 2 of firm i in year t. βnFE controls the firm effects. Ɛit is the error term for firm i in year t. β CONTROLS it represents the control variables of table 2 of firm i in year t. βnFE controls the firm effects. Ɛit is the error term for firm i in year t.
I expect a negative relation between the political directors and financial performance. Also, I expect a negative relation for senior bureaucrat directors and a positive relation for minister directors.
12
expect a negative relation for senior bureaucrat directors and a positive relation for minister directors. Table 2 shows the measurements of the variables. To mitigate bias by outliers, the dependent and control variables are all winsorized on a level of 1%.
4. RESULTS
4.1. Descriptive statistics
Table 3 shows the descriptive statistics of the dependent, independent, interacting and the control variables. The dependent, interacting and the control variables are winsorized at the bottom and top to a 1% level. The mean of ROA is 0.0088 with a standard deviation of 0.195. The mean of SALES_TA is 0.9732 with a standard deviation of 0.7733. The mean of
PERCENTAGE_POLITICAL_DIRECTORS is 0.076 with a standard deviation of 0.117. This means that 7.6% of the firms in our sample has a former politicians on their board. This is divided into former senior bureaucrats (PERCENTAGE_BUREAUCRAT_DIRECTORS) with a mean of 0.067 and a standard deviation of 0.104) and into former ministers (PERCENTAGE_MINISTER_DIRECTORS with a mean of 0.0067 and a standard deviation of 0.031). So, 7.6% of the firms in the sample have former politicians in their board from what 6.7% are former senior bureaucrats and 0.67% are former ministers.
[insert table 3]
4.2. Person Correlations Coefficient
In table 4 are the result of the Pearson correlations tests. This test is executed to determine what the correlation is between the variables and whether multicollinearity exists. I mention the correlation above the 0.7, because from this mark there is a possibility for multicollinearity. The highest correlation is between PERCENTAGE_BUREAUCRAT_DIRECTORS and
PERCENTAGE_POLITICAL_DIRECTORS of 0.952, because the senior bureaucrats are also included into the politicians in the board.
[insert table 4]
4.3. Regression
This study examines if political directors and firm performance relates and the moderate effect of successful firms. To test the three hypotheses a fixed-effect (within) regression is performed. In table 5 the results of the hypothesis 1 and 2 is presented and in table 6 the third hypothesis is described. Panel A of table 5 presents the direct relationship between political directors and financial
performance measured by the ROA and the SALES_PER_ASSET. Panel A of table 5 shows no significant relations for the measures for the hypothesis. The percentage of political directors
13
(PERCENTAGE POLITICAL DIRECTORS) has a negative relation with the ROA and is not significant (β=-0.009). Further, the Sales per asset (β=-0.100) is also negative and not significant. According to the theory the direction of this relationship is correct. Mentioned in the chapter 2, the relationship of politicians in the board on firm performance is negative. The results show that the TOBINS_Q is statistically significant associated with the ROA (β=0.006; p<0,005). The second significant result is the TOTAL_LIABITIES_PER_ASSET, this is positively significant associated with the ROA
(β=0.107; p<0,01). Furthermore, indicates that LOSS_DUMMY is negative significant associated with the ROA (β= -0.064; p<0,01). BRD_SIZE is negative significant associated with the ROA (β=-0.025; p<0,1). The TOTAL_LIABILTIES_PER_ASSET is positively significant associated with the SALES_PER_ASSET (β=0.581; p<0,01). BRD_SIZE is negative significant associated with the SALES_PER_ASSET (β=-0.122; p<0,01).
Panel B of table 5 presents the direct relationship between senior bureaucrat directors, minister directors and financial performance measured by the ROA and the SALES_PER_ASSET. Panel B of table 5 shows for senior bureaucrats a significant relation (SALES_PER_ASSET) (β=-0.163; p<0,01), the ROA is not significant (β=-0.020). The negative direction is in line with the theory in chapter 2. Because I posit that senior bureaucrat directors are likely to have a negative relation with financial performance. The results show that the Tobins’Q is (β=0.006; p<0,05) statistically significant associated with the ROA. Further, that the LIABILITES is significant (β=0.107; p<0,01) with the ROA. The LIABILITIES is also statistically significant with the SALES_PER_ASSET (β=0.580; p<0,01).
For ministers the relations are not significant, the direction of both variables is positive. For ROA (β=0.081) and for SALES_PER_ASSET (β=0.262). Statistically significant relations with the ROA is the R&D expenditure (β=-0.315; p<0,01) and the board size (β=-0.025; p<0,1). Further, the
TOTAL_LIABLITIES_PER_ASSET (β=0.578; p<0,01) and the board size (β=-0.115; p<0,01) have significant relations with the SALES_PER_ASSET.
[insert table 5]
For the third hypothesis the results are presented in Table 6. Table 6 panel A presents the
relationship of political directors in successful firms with financial performance measured by the ROA and the sales per asset. The direct relations between the senior bureaucrat directors in successful firms and financial performance measured by the ROA (β=-0.093; p<0,05) and the
SALES_PER_ASSET (β=-0.236; p<0,05) are statistically significant. The interacting results of senior bureaucrat directors in successful firms is positive significant associated with ROA (β=0.046;
p<0,05) and SALES_PER_ASSET (β=0.074; p<0,05). Accumulating the results of the direct and interacting results is the relationship between political connected board members in successful firm
14
with firm performance still negative. That is in line with the theory in chapter 2. Furthermore, the LIABILITIES is negative significant associated with ROA (β=-0.106; p<0,01). BOARD_SIZE is negative significant associated with the ROA (β=-0.025; p<0,1).
Panel B of table 6 presents the relationship between senior bureaucrat directors, minister directors in successful firms and financial performance measured by the ROA and the SALES_PER_ASSET. The direct relations between senior bureaucrat directors and the ROA (β=-0.122; p<0,01) and
SALES_PER_ASSET (β=-0.345; p<0,01) are negative significant. The interacting relation of senior bureaucrat directors in successful firms is positive significant associated with the ROA (β=0.057; p<0,01) and SALES_PER_ASSET (β=0.102; p<0,01). Accumulating the results of the direct and interacting results is the relationship between political connected board members in successful firm with firm performance still negative. This is in line with the second hypothesis. Furthermore, the LIABITIES is positive significant associated with the ROA (β=0.107; p<0,01). BOARD_SIZE is
negative significant associated with the ROA (β=-0.024; p<0,1). The LIABITIES is positive significant associated with the SALES_PER_ASSET (β=0.580; p<0,01). BOARD_SIZE is negative significant associated with the SALES_PER_ASSET (β=-0.110; p<0,1).
For ministers, the direct and moderate relations are not significant, the direction of both variables is positive. The direct relation with the ROA (β= 0.036) and SALES_PER_ASSET (β=0.253). The interacting relation of ministers in the board with successful firms is positive associated with the ROA (β=0.028) and for the SALES_PER_ASSET (β=0.006). BOARD_SIZE is negative significant
associated with the ROA (β=-0.025; p<0,1). The TOTAL_LIABILTIES_PER_ASSET is positive significant associated with the SALES_PER_ASSET (β=0.578; p<0.01). BOARD_SIZE is negative significant associated with the SALES_PER_ASSET (β=-0.115; p<0,1).
[insert table 6]
4.4. Robustness test
The results in the previous section suggest that political directors have a negative relation with financial performance. Second, it seems that senior bureaucrats have a negative relation with financial performance. Lastly, that successful firms are likely to moderate these negative
relationships. To strength the results hypotheses one, two and three I perform according to Ye et al. (2019) a two-stage least square (2SLS) to solve the endogeneity problems of the fixed-effect
regression. Through omitted variables the conclusion can be influenced by unobserved variables. By creating an instrumental variable is made to assess the industry average of firms. This table is unreported.
15
5.
CONCLUSION
This study investigated the relationship between political directors on the board and financial performance in the UK. I found that having political directors on the board is likely negatively
associated with performance. So, my results support my hypothesis stating that political directors are negatively associated with performance. Since political directors are unlikely to bring valuable
resources for performance to the firms in the UK. Though they bring their network and knowledge about the government to the firm, these resources are likely not beneficial in the UK, because it is a developed country.
I also found that having senior bureaucrats in the board is negatively associated with performance. Based on the resource dependency theory, I argue that more political knowledge and tenure has a negative impact on financial performance. I test this by making a distinction between senior
bureaucrats and ministers and their association with financial performance. Minister directors do not have a significant association with firm performance. My results support my hypothesis that senior bureaucrats have a negative association with financial performance.
Lastly, I found that successful firms moderate the negative association between political directors and financial performance. Successful firms can make better decision that resulted in high firm
performance and success. They hence seem to have more capabilities to make the right choices in uncertain situations. The results corresponding to all hypotheses based on fixed-effects regression. By performing the two-stage-least square regression as additional test, I strengthen the results of the fixed-effects regression. These results are significant and consistent with the fixed-effects regression. This thesis has some managerial and future research implications. The results show that in the UK political directors are likely negatively associated with financial performance. Furthermore, senior bureaucrats have a stronger political background, but are more negatively associated with
performance than ministers. The managerial implication is that firms may need to reconsider their practice of appointing political directors in boards. To strengthen this argument, I advocate
additional research assessing why political directors are appointed in the UK.
This thesis has some limitations for practice. This thesis focused only on the UK, therefore the
findings are not trivially generalisable to other countries. Second, this study focused on non-financial and non-utilities. This means that results may not be generalizable to other industries.
Further research could more closely focus on reason why political directors are appointed in the UK. Especially why senior bureaucrat directors are appointed.
16
6.
Literature
Adams, R. & Ferreira, D., 2007. A Theory of Friendly Boards. The Journal of Finance, 62(1), 217-250. Aggarwal, R.K., Felix, M, & Wang, T., 2012 Corporate political contributions: investment or agency? http:// ssrn.com/abstract=972670
Agrawal, A. & Knoeber, C., 2001. Do Some Outside Directors Play a Political Role? The Journal of Law and Economics, 44(1), 79-198.
Beckman, L. 2006. The Competent Cabinet? Ministers in Sweden and the Problem of Competence and Democracy. Scandinavian Political Studies, 29(2), 111-129.
Boubakri, N., Cosset, J. & Saffar, W., 2008. Political connections of newly privatized firms. Journal of Corporate Finance, 14(5), 654-673.
Blom-Hansen, J., Baekgaard, M., Christensen, J. & Serritzlew, S., 2017. Politicians and Bureaucrats: Reassessing the Power Potential of the Bureaucracy. Presentation NORKOM
Chari, A., Chen, W. & Dominguez, K., 2009. Foreign Ownership and Firm Performance: Emerging-Market Acquisitions in the United States. SSRN Electronic Journal.
Chen, J., Leung, W. & Evans, K., 2018. Female board representation, corporate innovation and firm performance. Journal of Empirical Finance, 48, 236-254.
Chu, J., 2019. Accruals, Investment, and Future Performance. Abacus, 55(4), 783-809.
Dunsire, A., 1988. Bureaucratic Morality in the United Kingdom. International Political Science Review, 9(3), 179-191.
Fama, E., & Jensen, M., 1983. Separation of ownership and control. Journal of Law and Economics, 26(2), 301–325.
Faccio, M., 2010. Differences between Politically Connected and Nonconnected Firms: A Cross-Country Analysis. Financial Management, 39(3), 905-928.
Faccio, M., Masulis, R. & McConnell, J., 2006. Political Connections and Corporate Bailouts. The Journal of Finance, 61(6), 2597-2635.
Fauver, L., Hung, M., Li, X. & Taboada, A., 2017. Board reforms and firm value: Worldwide evidence. Journal of Financial Economics, 125(1), 120-142.
17
Fisman, R., 2001. Estimating the Value of Political Connections. American Economic Review, 91(4), 1095-1102.
Garrett, A., 2019. A Comparison of United Kingdom and United States Approaches to Board Structure. The Corporate Governance Law Review, 3(2), 93-114.
González-Bailon, S., Jennings, W. & Lodge, M., 2012. Politics in the Boardroom: Corporate Pay, Networks and Recruitment of Former Parliamentarians, Ministers and Civil Servants in
Britain. Political Studies, 61(4), 850-873.
Guest, P., 2008. The determinants of board size and composition: Evidence from the UK. Journal of Corporate Finance, 14(1), 51-72.
Guest, P., 2009. The impact of board size on firm performance: evidence from the UK. The European Journal of Finance, 15(4), 385-404
Hadani, M. & Schuler, D., 2012. In search of El Dorado: The elusive financial returns on corporate political investments. Strategic Management Journal, 34(2), 165-181.
Herciu, M. & Ogrean, C., 2017. Does Capital Structure Influence Company Profitability? Studies in Business and Economics, 12(3), 50-62.
Hillman, A., 2005. Politicians on the Board of Directors: Do Connections Affect the Bottom Line? Journal of Management, 31(3), 464-481.
Hillman, A. & Hitt, M., 1999. Corporate Political Strategy Formulation: A Model of Approach, Participation, and Strategy Decisions. The Academy of Management Review, 24(4), 825. Ireland, R. & Miller, C., (2004). Decision-making and firm success. Academy of Management Perspectives, 18(4), 8-12.
Jensen, M. & Meckling, W., 1976. Theory of the firm: managerial behavior, agency cost, and ownership structure. Journal of Financial Economics, 3, 305–360.
Kim, K., Mauldin, E. & Patro, S. 2014. Outside directors and board advising and monitoring performance. Journal of Accounting and Economics, 57(2-3), 110-131.
Kor, Y. & Sundaramurthy, C., 2008. Experience-Based Human Capital and Social Capital of Outside Directors. SSRN Electronic Journal.
Laffont, J. & Mordimort. D., 2002. The Theory of Incentives: The Principal-Agent Model. Princeton: Princeton University Press.
18
La Porta, R., Lopez‐ de‐ Silanes, F., Shleifer, A. & Vishny, R., 1998. Law and Finance. Journal of Political Economy, 106(6), 1113-1155.
Mitchell, J., 2009. Where are we now? The UK recession and nowcasting GDP growth using statistical models. National Institute Economic Review, 209(1), 60-69.
Mura, R., 2007. Firm Performance: Do Non-Executive Directors Have Minds of their Own? Evidence from UK Panel Data. Financial Management, 36(3), 81-112.
Niessen, A. & Ruenzi, S., 2009. Political Connectedness and Firm Performance: Evidence from Germany. German Economic Review, 11(4), 441-464.
Ozkan, N., 2011. CEO Compensation and Firm Performance: an Empirical Investigation of UK Panel Data. European Financial Management, 17(2), 260-285.
Otley, D., 1999. Performance management: a framework for management control systems research. Management Accounting Research, 10(4), 363-382.
Pfeffer, J. & Salancik, G., 1978. The external control of organizations: a resource dependence perspective. Stanford, Calif.: Stanford Business Books.
Ruigrok, W., Peck, S., Tacheva, S., Greve, P. & Hu, Y., 2006. The Determinants and Effects of Board Nomination Committees*. Journal of Management & Governance, 10(2), 119-148.
Sapienza, P., 2004. The effects of government ownership on bank lending. Journal of Financial Economics, 72(2), 357-384.
Shaukat, A. & Trojanowski, G., 2018. Board governance and corporate performance. Journal of Business Financial Accounting, 45, 184–208.
Sher, P. & Yang, P., 2005. The effects of innovative capabilities and R&D clustering on firm performance: the evidence of Taiwan's semiconductor industry. Technovation, 25(1), 33-43.
Shleifer, A, & Vishny, R., 1997. A survey of corporate governance. Journal of Finance, 52, 737–783. Su, Z. & Fung, H., 2013. Political Connections and Firm Performance in Chinese Companies. Pacific Economic Review, 18(3), 283-317.
Thomsen, S., 2003. The Convergence of Corporate Governance Systems to European and Anglo-American Standards. European Business Organization Law Review, 4(1), 31-50.
19
Wood, B., 1988. Principals, Bureaucrats, and Responsiveness in Clean Air Enforcements. American Political Science Review, 82(1), 213-2348
Ye, D., Deng, J., Liu, Y., Szewczyk, S. & Chen, X., 2019. Does board gender diversity increase dividend payouts? Analysis of global evidence. Journal of Corporate Finance, 58, 1-26.
20
7.
Tables
Table 1: Sample statistics
Year Observations 2002 285 2003 317 2004 363 2005 467 2006 573 2007 685 2008 699 2009 703 2010 612 2011 644 2012 649 2013 594 2014 553 2015 561 7.705
21
Table 2: Variable measurements
Variable Measurement Source
Return on assets Measure of accounting performance: operational income / total assets BoardEx, Worldscope
Sales per asset Measure of accounting performance: revenues / total assets BoardEx, Worldscope
Tobins'Q * Measure of firm value: (Total assets - Common shareholders equity + Market cap) / Total assets Worldscope
Advisor director At least one director on the board with former governmental advisory employment Hand-collected / BoardEx
Senior bureaucrat director At least one director on the board with former senior bureaucratic employment Hand-collected / BoardEx
Minister director At least one director on the board with former minister employment Hand-collected / BoardEx
Leverage* Percentage of leverage: Total liabilities / total assets BoardEx, Worldscope
loss 1 if if roa < 0; 0 otherwise BoardEx, Worldscope
Capital expenditures/ total
assets* Percentage of capital expenditure in relation with the total assets: capital expenditure/ total assets BoardEx, Worldscope
R&D / total assets* rnd/assets_total (0 if rnd_ta.) BoardEx, Worldscope
Closely held shares* Measure of ownership: amount of closely held shares / 100 BoardEx, Worldscope
Foreign assets / total assets* Percentage of assets which are foreign: Foreign / total assets BoardEx, Worldscope
Board size* Total number of Board of directors BoardEx, Worldscope
Board independence* Percentage of non-executive directors on the board: non-executive directors / board size BoardEx, Worldscope
Female director* Percentage of female directors / board size BoardEx, Worldscope
Foreign board directors* Percentage of foreign directors / board size BoardEx, Worldscope
Board tenure* Percentage of director board tenure / board size BoardEx, Worldscope
Quoted board affiliations* Total years membership boards of quoted companies/Board size BoardEx, Woldscope
This table shows the measurements of the dependent, independent, interaction and lastly the control variables used in this study. When * stands after the variable, this means that the variable is winsorized at 1% level at the top and bottom.
22
Table 3: Descriptive Statistics Variable Obs Mean Std. Dev. Min Max
ROA 7,705 .009 .195 -0,863 .352
Sales per asset 7,705 .973 .773 0 3.620
Percentage political directors 7,705 .076 .117 0 .75
Percentage bureaucrat directors 7,705 .067 .105 0 .6
Percentage minister directors 7,705 .007 .032 0 .333
Tobins'Q 7,705 1,738 1.337 0 9,555
Total liabilities/ Total assets 7,705 .456 .229 0,008 1
Loss 7,705 .309 .462 0 1
Capital expenditures / Total assets 7,705 .045 .058 0 .350
R&D/ Total assets 7,705 .028 .081 0 .753
Closesly held shares 7,705 .317 .247 0 .934
Foreign assets / Total assets 7,705 .247 .315 0 1
Board size 7,705 1,862 .336 1,099 2.944
Board independence 7,705 .495 .177 0 1
Female director 7,705 .065 .099 0 .429
Foreign board 7,705 .168 .230 0 .857
Board tenure 7,705 .924 .671 0 3.536
Quoted board affiliations 7,705 .274 .133 0 .875
This table shows the descriptive statistics of the variables used in this study. The variables are winsorized at the bottom and top at a 1% level. The explanation of the variables is presented in table 2. The dependent variables (ROA and SALES_PER_ASSET) are used in 8.8% and 97.3% as measures in the financial statement, by the firms in this sample. Further that 7.6% of the firms in our sample have a political directors on their board. This is divided into bureaucrat directors and minister directors. So, 7.6% of the firms in the sample have former politicians in their board from what 6.7% are former senior bureaucrats and 0.67% are former ministers.
23
Table 4: Correlation matrix
Variables 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
1 ROA ---
2 Sales per asset 0.315 ---
3 Percentage political directors 0.0875 0.0353 ---
4 Percentage bureaucrat directors 0.0979 0.0406 0.952 ---
5 Percentage minister directors -0.0173 -0.00494 0.339 0.0776 ---
6 Tobins'Q -0.272 -0.0234 0.0190 0.00823 0.0270 ---
7 Total liabilities/ Total assets 0.241 0.451 0.151 0.148 0.0510 -0.0846 ---
8 Loss -0.654 -0.295 -0.0894 -0.108 0.0333 0.108 -0.252 ---
9 Capital expenditures / Total assets 0.0615 -0.0803 0.0170 0.00543 0.0318 0.0300 0.0187 -0.0243 ---
10 R&D/ Total assets -0.463 -0.103 0.0235 0.0275* 0.00110 0.370 -0.139 0.223 -0.0930 ---
11 Closesly held shares -0.124 -0.0135 -0.178 -0.188 -0.0213 -0.0241 -0.161 0.155 0.0139 0.00743 ---
12 Foreign assets / Total assets 0.0596 -0.216 0.0422 0.0419 0.00389 -0.0837 -0.0681 0.0362 0.136 -0.109 -0.0969 ---
13 Board size 0.262 0.0455 0.292 0.295 0.0553 -0.00786 0.255 -0.265 0.0593 -0.0486 -0.276 0.123 ---
14 Board independence 0.115 -0.0851 0.231 0.234 0.0344 -0.0163 0.0736 -0.0692 0.0324 -0.0235 -0.232 0.215 0.315 ---
15 Female director 0.110 0.0851 0.150 0.154 0.0206 0.0211 0.127 -0.102 -0.0228 -0.0596 -0.103 -0.0325 0.196 0.163 ---
16 Foreign board -0.0244 -0.252 0.0626 0.0600 0.00697 0.00702 -0.109 0.0726 0.0927 -0.0587 0.00444 0.416 0.172 0.309 0.0642 ---
17 Board tenure 0.0453 0.108 -0.180 -0.173 -0.0667 -0.0603 -0.110 -0.0810 -0.0418 0.00446 0.195 -0.127 -0.465 -0.261 -0.113 -0.188 ---
18 Quoted board affilations -0.165 -0.135 -0.0465 -0.0501 0.00782 -0.00301 -0.191 0.209 -0.0419 -0.0322 0.0556 0.0744 -0.601 0.0252 -0,092 0.0726 0.170 ---
This table presents the Pearson Correlation Coefficients of the variables where numbers in bold have at least a p<0.05 significance level. This test is performed to determine of multicollinearity exists between the variables. The highest correlation is PERC_BUR_DIR of 0.952, because the senior bureaucrats are also included into the politicians in the board. This is according to table 1 the largest represented in the sample.
24
Table 5: Fixed effect regression analysis
Panel A Panel B
VARIABLES ROA Sales per asset ROA Sales per asset ROA Sales per asset
Percentage political directors -0.009 -0.100
(0.026) (0.083)
Percentage bureaucrat directors -0.020 -0.163*
(0.030) (0.088)
Percentage minister directors 0.081 0.262
(0.077) (0.239)
Tobins'Q 0.006** -0.004 0.006** -0.004 0.006** -0.004
(0.003) (0.006) (0.003) (0.006) (0.003) (0.006)
Total liabilities/ Total assets 0.107*** 0.581*** 0.107*** 0.580*** 0.106*** 0.578***
(0.019) (0.050) (0.019) (0.050) (0.019) (0.051)
Loss -0.064*** -0.022 -0.064*** -0.022 -0.064*** -0.022
(0.005) (0.015) (0.005) (0.015) (0.005) (0.015)
Capital expenditures / Total assets -0.011 -0.178* -0.010 -0.178* -0.011 -0.181*
(0.043) (0.105) (0.043) (0.105) (0.044) (0.104)
R&D/ Total assets -0.315*** 0.039 -0.315*** 0.041 -0.315*** 0.040
(0.092) (0.139) (0.092) (0.139) (0.092) (0.140)
Closesly held shares -0.008 0.031 -0.008 0.031 -0.008 0.029
(0.013) (0.036) (0.013) (0.036) (0.013) (0.036)
Foreign assets / Total assets -0.003 -0.005 -0.003 -0.005 -0.003 -0.003
(0.010) (0.031) (0.010) (0.031) (0.010) (0.030) Board size -0.025* -0.112*** -0.025* -0.111*** -0.025* -0.115*** (0.014) (0.040) (0.014) (0.040) (0.014) (0.040) Board independence 0.006 -0.003 0.006 -0.001 0.005 -0.008 (0.020) (0.051) (0.020) (0.051) (0.020) (0.051) Female director 0.013 0.001 0.014 0.005 0.012 -0.006 (0.026) (0.085) (0.026) (0.086) (0.026) (0.084) Foreign board 0.036* 0.016 0.036* 0.015 0.036* 0.018 (0.020) (0.058) (0.020) (0.058) (0.020) (0.058) Board tenure 0.017** -0.014 0.017** -0.014 0.017** -0.013 (0.008) (0.016) (0.008) (0.016) (0.008) (0.016)
Quoted board affiliations -0.009 -0.220*** -0.008 -0.217*** -0.010 -0.228***
(0.026) (0.073) (0.026) (0.073) (0.026) (0.073)
Constant -0.014 1.109*** -0.014 1.109*** -0.014 1.110***
(0.039) (0.096) (0.039) (0.096) (0.039) (0.096)
Firm fixed effects Yes Yes Yes Yes Yes Yes
Year fixed effects Yes Yes Yes Yes Yes Yes
Observations 7,705 7,709 7,705 7,709 7,705 7,709
R-squared 0.087 0.075 0.087 0.076 0.087 0.075
Number of firm_id 1,345 1,347 1,345 1,347 1,345 1,347
This table presents the results of the fixed effects model performed for the first and second hypothesis. The explanation of the variables used in this regression is in table 2. The results of panel A is that political directors are negatively associated with financial performance. This is measured with the ROA and the SALES_PER_ASSET. The results of panel B is that bureaucrat directors are negatively associated with financial performance. Lastly, that minister directors have a positive association with financial performance. The firm fixed effects and the year fixed effects are controlled in this regression. The robust standard errors are presented in parentheses. The presentation of significant correlation levels is *** p<0.01, ** p<0.05, * p<0.1.
25
Table 6: Fixed effect regression analysis
Panel A Panel B
VARIABLES ROA Sales per asset ROA Sales per asset ROA Sales per asset
Percentage political directors -0.093** -0.236**
(0.038) (0.099)
Percentage bureaucrat directors -0.122*** -0.345***
(0.040) (0.105)
Percentage minister directors 0.036 0.253
(0.139) (0.275)
Percentage political directors in successful firm 0.046** 0.074**
(0.020) (0.029)
Percentage bureaucrat directors in successful firm 0.057*** 0.102***
(0.022) (0.038)
Percentage minister directors in successful firm 0.028 0.006
(0.074) (0.074)
Tobins'Q 0.003 -0.009 0.003 -0.010 0.006** -0.004
(0.003) (0.007) (0.003) (0.007) (0.003) (0.006)
Total liabilities/ Total assets 0.106*** 0.580*** 0.107*** 0.580*** 0.106*** 0.578***
(0.019) (0.050) (0.019) (0.050) (0.019) (0.051)
Loss 0.064*** - -0.022 -0.064*** -0.022 0.065*** - -0.022
(0.005) (0.015) (0.005) (0.015) (0.005) (0.015)
Capital expenditures / Total assets -0.013 -0.183* -0.013 -0.183* -0.011 -0.181*
(0.043) (0.105) (0.043) (0.104) (0.044) (0.104)
R&D/ Total assets 0.320*** - 0.032 -0.317*** 0.037 0.317*** - 0.040
(0.092) (0.139) (0.092) (0.139) (0.092) (0.141)
Closesly held shares -0.007 0.032 -0.007 0.032 -0.008 0.029
(0.013) (0.036) (0.013) (0.036) (0.013) (0.036)
Foreign assets / Total assets -0.003 -0.005 -0.003 -0.005 -0.003 -0.003
(0.010) (0.030) (0.010) (0.030) (0.010) (0.030) Board size -0.025* -0.112*** -0.024* -0.110*** -0.025* -0.115*** (0.014) (0.040) (0.014) (0.040) (0.014) (0.040) Board independence 0.006 -0.002 0.007 0.000 0.005 -0.008 (0.020) (0.051) (0.020) (0.051) (0.020) (0.051) Female director 0.008 -0.007 0.009 -0.004 0.012 -0.006 (0.026) (0.085) (0.026) (0.085) (0.026) (0.084) Foreign board 0.036* 0.017 0.036* 0.016 0.036* 0.018 (0.020) (0.058) (0.020) (0.058) (0.020) (0.058) Board tenure 0.016** -0.015 0.016** -0.015 0.017** -0.013 (0.008) (0.016) (0.008) (0.016) (0.008) (0.016)
Quoted board affiliations -0.010 -0.222*** -0.008 -0.217*** -0.010 -0.228***
(0.026) (0.073) (0.026) (0.073) (0.026) (0.073)
Constant -0.008 1.118*** -0.010 1.117*** -0.013 1.110***
(0.039) (0.096) (0.039) (0.096) (0.039) (0.096)
Firm fixed effects Yes Yes Yes Yes Yes Yes
Year fixed effects Yes Yes Yes Yes Yes Yes
Observations 7,705 7,709 7,705 7,709 7,705 7,709
R-squared 0.091 0.076 0.091 0.077 0.088 0.075
Number of firm_id 1,345 1,347 1,345 1,347 1,345 1,347
This table presents the results of the fixed effects model performed for the third hypothesis. The explanation of the variables used in this regression is in table 2. The results of panel A is that successful firms have a moderate effect on the negative relation between political
26
directors and financial performance. This is measured with the ROA and the SALES_PER_ASSET. The results of panel B is that successful firms have a moderating effect on the negative relation between bureaucrat directors and financial performance. Lastly, that successful firms strength the positive relation between minister directors and financial performance. The firm fixed effects and the year fixed effects are controlled in this regression. The robust standard errors are presented in parentheses. The presentation of significant correlation levels is *** p<0.01, ** p<0.05, * p<0.1.
